[vorbis-dev] Lossless compression

[Monty]

> The back-end encoder for Vorbis is codeword based;  an n-bit number is mapped
> to an m-bit number (or vector of numbers).  Compression works when n is
> generally less than m.

Eh, I was mixing my encoding and decoding metaphors.  Let me restate that,
keeping it all encode side:

The back-end encoder for Vorbis is codeword based;  an n-bit number or vector
of numbers is mapped to (encoded as) an m-bit codeword.  Compression works when
m is generally less than n.

Monty

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/